Written and maintained by CASRAI Editorial Board
Last updated
The Human Gene Mutation Database (HGMD) is an expert-curated repository of published germline mutations in human nuclear genes associated with inherited disease. Geneticists, clinical diagnosticians, and bioinformatics researchers use HGMD to verify whether a sequenced variant has been documented in peer-reviewed literature. Before querying genomic findings, investigators must understand how the database operates, what distinguishes its Public and Professional editions, and how its literature-driven variant classifications compare to clinical consensus standards.
What it is
HGMD constitutes an international reference archive of germline mutations associated with human inherited clinical conditions. Unlike broad public archives that host crowdsourced variant submissions, HGMD relies on manual curation by scientific experts who screen biomedical journals to index published mutational events. Each entry links a specific genetic lesion to its original report, capturing the disease phenotype, gene symbol, and observed mutational mechanism.
The database catalogs several molecular classes of nuclear variation:
- Single nucleotide substitutions: Missense, nonsense, and non-coding regulatory base changes.
- Splicing variants: Alterations within invariant splice junctions or regulatory splicing motifs.
- Micro-insertions and micro-deletions: Small indels under 20 base pairs causing frameshift or in-frame changes.
- Gross lesions: Large-scale structural alterations, including deletions, duplications, and inversions exceeding 20 base pairs.
- Repeat expansions: Pathogenic short tandem repeat expansions linked to inherited neurological disorders.
HGMD functions as an index of published evidence rather than an independent clinical consensus body. It provides researchers with direct citations to original papers, allowing rapid verification of historical clinical precedent.
Who runs it and what it covers
HGMD was established in 1996 by David N. Cooper and Michael Krawczak at Cardiff University’s Institute of Medical Genetics. Scientific curation continues at Cardiff University under Peter D. Stenson and colleagues. Commercial packaging and software distribution are managed by QIAGEN Digital Insights.
The database catalogs germline mutations in nuclear genes underlying inherited disorders. It excludes somatic mutations in tumors (cataloged by resources like The Cancer Genome Atlas (TCGA) and COSMIC) and mitochondrial DNA variants (tracked by MITOMAP).
Foundational peer-reviewed publications documented over 289,000 distinct lesions across 11,100 genes from 73,000 journal articles. Ongoing quarterly curation by QIAGEN Digital Insights has expanded the repository to more than 575,000 to 594,000 published germline mutations and polymorphisms.
HGMD assigns cataloged variants to five functional categories based on published literature claims:
| Tag | Category Name | Operational Definition |
|---|---|---|
| DM | Disease-causing mutation | Reported in published literature as directly responsible for human inherited disease. |
| DM? | Likely disease-causing mutation | Reported as disease-causing, but curators noted author uncertainty or disputed pathogenicity. |
| DP | Disease-associated polymorphism | Polymorphism statistically associated with altered disease susceptibility or phenotype modification. |
| DFP | Disease polymorphism with functional evidence | Polymorphism with experimental in vitro or in vivo evidence demonstrating altered function. |
| FP | Functional polymorphism | Polymorphism demonstrating biological effect, but lacking established disease association. |
Users must distinguish between the two available database editions. The public edition serves registered academic and non-profit researchers, whereas HGMD Professional is a licensed platform for diagnostic laboratories:
| Feature | HGMD Public | HGMD Professional |
|---|---|---|
| Updates | Maintains a 3-year publication lag | Updated quarterly with recent literature |
| Volume | Subset with 30% to 40% fewer entries | Complete collection of curated mutations |
| Coordinates | Restricted coordinate mapping | Full genomic coordinates (GRCh37/GRCh38) and HGVS |
| Search | Single-gene or accession queries | Multi-attribute search, mutation maps, and batch mode |
| Integration | Single-record web browsing only | VCF and flat-file downloads, APIs, and clinical tools |
| History | Static snapshot of public records | Full audit logs of variant re-evaluations |
How to search or use it
Querying HGMD involves seven systematic steps based on official platform workflow parameters:
- Access the portal: Open the authorized interface (Cardiff University academic portal or QIAGEN Digital Insights commercial application).
- Select search mode: Choose Gene-oriented, Mutation-oriented, Disease-oriented, Reference-oriented, or Batch Search.
- Enter standardized identifiers: Supply an HGNC gene symbol, an NCBI RefSeq accession (e.g., NM_000000), or a dbSNP rsID. Obsolete aliases should be avoided.
- Specify variant coordinates: In mutation-oriented queries, enter the cDNA nucleotide position, amino acid substitution, or genomic coordinates (GRCh37 or GRCh38).
- Filter by classification: Narrow hits by mutational category (missense, regulatory, splicing, or indel) and variant tag (such as DM or DM?).
- Review mutation records: Open the entry to inspect the affected codon, exon/intron boundary, reported phenotype, and primary publication citation.
- Evaluate literature and export: Check cited publications (via PMID/DOI) to assess family segregation and study design. Licensed users can export VCF records for local bioinformatic pipelines.
Strengths and limits
Understanding HGMD’s strengths and operational constraints is necessary for accurate genomic interpretation.
Key Strengths
- Historical depth: More than two decades of manual curation capturing rare mutations published prior to high-throughput sequencing.
- Direct attribution: Every entry links directly to the peer-reviewed report that identified the mutation.
- Phenotypic context: Records include specific disease names, assisting in rare disease differential diagnosis.
- Coordinate standardization: HGMD Professional converts historic literature descriptions into modern GRCh37 and GRCh38 genomic coordinates.
Operational Limitations
- Literature attribution bias: Because HGMD records published assertions, historic entries may reflect obsolete genetic criteria.
- Legacy false-positive rates: Benchmark studies indicate that some older literature-reported DM variants show high allele frequencies in healthy cohorts within gnomAD (Genome Aggregation Database), indicating benign population variation.
- Exclusion of benign variation: HGMD does not systematically collect benign or likely benign variants; it cannot serve as a standalone negative filtering tool.
- Conservative reclassification: Because HGMD indexes published claims, entries are rarely downgraded unless formal refutations appear in print.
Alternatives
Clinical laboratories routinely cross-reference literature databases with clinical assertion archives, population allele catalogs, in silico algorithms like PolyPhen-2, and orthogonal validation via Sanger sequencing.
| Resource | Primary Scope | Curation Model | Access Model | Key Difference vs. HGMD |
|---|---|---|---|---|
| ClinVar (NCBI) | Clinical variant assertions | Crowdsourced laboratory submissions (ACMG/AMP) | Free public domain | Catalogs benign to pathogenic variants; updates rapidly as laboratories submit evidence. |
| gnomAD | Population allele frequencies | Standardized processing of exome/genome cohorts | Free public domain | Provides unselected baseline frequencies to filter out common benign variants. |
| ClinGen | Gene and variant curation | Expert panel consensus under clinical frameworks | Free public domain | Evaluates clinical validity for high-priority genes rather than indexing all literature. |
| LOVD | Locus-specific gene variations | Community-driven curation by gene stewards | Free open access | Organized around locus databases; curation depth varies by gene community. |
| COSMIC | Somatic mutations in cancer | Manual literature curation of tumors | Academic free / Commercial license | Focuses on acquired tumor mutations; HGMD concentrates on inherited germline conditions. |
FAQ
What is the primary difference between HGMD and ClinVar?
HGMD is a curated literature index cataloging published claims of disease-causing and functional germline variants. ClinVar is an open-access NCBI repository where diagnostic laboratories and expert panels deposit variant interpretations classified across the full ACMG/AMP spectrum (from benign to pathogenic).
What does the DM classification indicate in HGMD?
The DM designation signifies a disease-causing mutation. It indicates that the authors of the cited study reported the variant as directly responsible for the patient’s inherited condition, representing the literature claim rather than an independent ACMG/AMP consensus evaluation.
Is HGMD available for free?
A public edition is accessible to registered academic and non-profit researchers via Cardiff University. However, it operates with a 3-year publication lag, contains 30% to 40% fewer entries, and lacks batch querying and full coordinate downloads. HGMD Professional requires a commercial license from QIAGEN Digital Insights.
Does HGMD catalog somatic cancer mutations?
No. HGMD catalogs germline variants in nuclear genes associated with inherited disease. Somatic alterations in neoplastic tissues are cataloged by dedicated oncology databases such as COSMIC and TCGA.
How frequently is HGMD updated?
HGMD Professional is updated quarterly by QIAGEN Digital Insights with newly published variants and refined coordinates. The public edition updates less frequently and maintains a deliberate 3-year publication delay.
Sources
- QIAGEN Digital Insights. Human Gene Mutation Database (HGMD®) Professional Documentation. Available at: https://digitalinsights.qiagen.com/products-overview/clinical-insights-portfolio/human-gene-mutation-database/. Accessed November 9, 2026.
- Stenson, P. D., Mort, M., Ball, E. V., Chapman, M., Evans, K., Azevedo, L., Hayden, M., Heywood, S., Millar, D. S., Phillips, A. D., and Cooper, D. N. (2020). The Human Gene Mutation Database (HGMD®): optimizing its use in a clinical diagnostic or research setting. Human Genetics, 139(10), 1197–1207. PMID: 32596782; PMCID: PMC7497289; DOI: 10.1007/s00439-020-02199-3. Accessed November 9, 2026.
- Cardiff University Institute of Medical Genetics. The Human Gene Mutation Database. Available at: https://www.hgmd.cf.ac.uk/. Accessed November 9, 2026.
- National Center for Biotechnology Information (NCBI). ClinVar Overview. Available at: https://www.ncbi.nlm.nih.gov/clinvar/intro/. Accessed November 9, 2026.
